ABSTRACT Diffuse large B-cell lymphoma (DLBCL) is the most common and a clinically aggressive lymphoma. In a genome-wide association study (GWAS) of DLBCL, we identified and validated the first genome-wide significant loci for persons of European ancestry at 6p21, 6p25, 8q24 and 2p23. However, the specific functional variant(s) of the GWAS-discovered loci have not been identified, so a next critical step is to fine-map these regions and conduct a bioinformatics analysis to characterize potential genetic drivers. Furthermore, DLBCL is biologically and clinically heterogeneous, with this heterogeneity in part defined by cell-of-origin (COO) and MYC status. COO derives from gene expression studies and has two dominant subtypes ? germinal center (GCB) and nonGCB. In preliminary data from our immunogenetic studies, SNPs from 6p21 were strongly associated with follicular lymphoma (FL), another germinal center lymphoma, and these same SNPs were associated with GCB but not with nonGCB DLBCL. This suggests a shared genetic etiology for FL and GCB-DLBCL at least for some MHC loci. MYC dysregulation through MYC rearrangements, particularly in concert with BCL2 and/or BCL6 rearrangements (?double/triple hit?), as well as aberrant MYC expression are associated with aggressive DLBCL; whether there is heterogeneity of the germline risk variants, particularly at the 8q24 region (location of MYC), by MYC status is unknown. Finally, patients with FL can later develop DLBCL (transformation), which is highly aggressive. We also found that the same SNPs from 6p21 were also associated with an increased risk of FL transformation to DLBCL, which provides new etiologic insights into de novo DLBCL. Comprehensive follow-up of these new and compelling findings provide the rationale and overall goals of our application. Our aims are: (1) To characterize the newly discovered DLBCL GWAS loci; (2) To evaluate etiologic heterogeneity of genetic risk for DLBCL molecular subtypes; and (3) To evaluate the role of germline genetic variants and tumor markers with risk of transformation from FL to DLBCL. To meet our aims, we will use the existing and ongoing resources of the Mayo Case-Control Study and the Iowa-Mayo SPORE and our established collaborations with InterLymph, MD Anderson, Emory University, LYSA (French Lymphoma Trials Group) and ECOG (Eastern Oncology Group). This proposal is a logical and critical next step to follow-up our novel DLBCL GWAS loci, and will provide new insights into the genetic architecture of risk for DLBCL, DLBCL molecular subtypes, and FL transformation. DLBCL and FL are the two most common lymphoma subtypes, and FL transformation is an important clinical problem. At the completion of this project, we expect to have defined the location of risk SNPs for DLBCL and its molecular subtypes. Further, we will have provided unique insights into the shared pathogenesis of FL and GCB-DLBCL, as well as FL transformation to DLBCL. Collectively, our findings should have a major impact on our understanding of DLBCL pathobiology to inform etiologic mechanisms, risk assessment, prevention and treatment. 1